JPH03200061A - Ultrasonic flaw detecting method - Google Patents

Abstract

PURPOSE:To improve flaw detection precision close to the end of the measuring surface of a test material by obtaining a correction factor to exclude the effects of the ineffective sound wave and multiple reflection with a reference material and correcting the measured flaw detection data with the factor. CONSTITUTION:A flaw detection test is firstly made on the reference material. The flaw detection data for each test position are stored in a reference material flaw detection data memory 10. After the tests on all the test positions are finished, the average value of the data is obtained by an average value arithmetic means 11 at the position where the ineffective sound wave and multiple reflection are not caused. The correction factor for each test position is then obtained by he correction factor for each test position is then obtained by the correction factor arithmetic means 12 from a specified equation based on the average value and the data for each test position. The correction factor obtained by the memory 12 is stored in a correction factor memory 13. A flaw detection test is then made on the test material. The measured data at each test position are stored in a test material flaw detection data memory 14. The measured data stored in the memory 14 are multiplied by the corresponding correction factor at each test position stored in the memory 13 by a data correction arithmetic means 15, and the corrected data are indicated 16.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an ultrasonic flaw detection method for detecting flaws in a test material by a water immersion method.

[Prior art] Typical ultrasonic flaw detection methods include the direct contact method, in which a probe is brought into direct contact with a test specimen, and the water immersion method, in which a test material is tested in water. There is. The water immersion method is
It is especially suitable for high-speed automatic flaw detection, and is often used to inspect materials and processed parts. FIG. 8 shows an outline of a general apparatus for performing a flaw detection test using the water immersion method. In the figure, 1 is a container containing liquid, 2 is a pair of materials 2
A test material made by joining a and 2b, 3 a probe, 4 an electronic pulse is given to the probe to generate ultrasonic waves from the probe, and the reflected echo reflected from the test material is detected. It is an ultrasonic pulser-receiver that converts into amplified electronic pulses. 5 is an A/D converter that converts the signal output from the ultrasonic pulser/receiver 4 into a digital signal, and 6 is a microcomputer that processes the digital signal representing the reflected echo and displays the result on the cathode ray tube 7. .

The display mode on the cathode ray tube 7 is arbitrary.

The probe 3 includes a focus type probe that can narrow the focus, and a flat type probe that emits ultrasonic waves without narrowing the focus.

[Problem to be solved by the invention] No matter which of the above probes is used to perform the flaw detection test,
As the thickness of the test material increases, there is a problem in that the accuracy of flaw detection near the ends of the test material decreases. FIG. 9 shows changes in the state of ultrasonic waves and reflected echoes when a focus type probe 3 is used to perform a flaw detection test on the state of a bonded portion of a test material 2 in which a pair of materials are bonded. In this figure, only one material 2a constituting the test material 2 is shown. Note that the reference numeral 8 is a surface on which a flaw detection test is performed, that is, a measurement surface, and in this example corresponds to the bonding surface of the materials 2a and 2b.

In FIG. 9, in the state (■), most of the ultrasonic waves output from the probe 3 are invalid waves that do not enter the test material 2, and the measurement surface 8 of the ultrasonic waves that have entered the test material 2 Since most of the reflected echoes reflected by the probe 3 are reflected in directions that cannot be detected by the probe 3, the reflected echoes cannot be detected sufficiently. Furthermore, even if the probe 3 moves and enters the state (■), half of the ultrasonic waves are still inactive. Even if the probe 3 moves further and enters the state shown in (2), a part of the ultrasonic waves becomes an invalid sound wave. In the state (2), it becomes possible to detect reflected echoes without being affected by the edge of the dovetail.

Therefore, even if flaw detection is performed based on the reflected echoes detected by the probe 3, flaws existing near the ends of the test material 2 cannot be detected with high accuracy.

FIG. 10 shows changes in the state of ultrasonic waves and reflected echoes when a flat type probe 3 is used. In this case, when the probe 3 is in the state of ■■, a part of the ultrasonic waves becomes an invalid sound wave. In addition, in the state of ■■, the ultrasonic wave that should spread outward and not return to the probe 3 undergoes multiple reflection (MR) and returns to the probe 3, so that the reflected echo becomes large. Then, in the state of ■■, it becomes possible to detect reflected echoes without being affected by almost invalid sound waves or multiple reflections.

Therefore, the reflected echo detected by the probe 3 cannot accurately represent the state of the measurement surface.

An object of the present invention is to provide an ultrasonic flaw detection method that can significantly improve flaw detection accuracy near the ends of a test material when performing a flaw detection test using a water immersion method.

[Means for Solving the Problems] In the invention of claim 1, first, the material is formed of the same material as the test material, and when overlapped, it has substantially the same shape as the test material, and the overlapping surface corresponds to the measurement surface of the test material. Prepare a pair of reference materials to be formed at the positions to be used.

Then, a flaw detection test is performed at each predetermined test position with the pair of reference materials superposed one on top of the other, and flaw detection data at each test position is measured. The average value of the flaw detection data at positions where no invalid sound waves and multiple reflections do not occur is determined, and a correction coefficient is determined from the ratio of the average value to the previously determined flaw detection data for each test position. Conduct a flaw detection test on the test material and obtain measured flaw detection data. Then, the actually measured flaw detection data is corrected by multiplying the actually measured flaw detection eta by a correction coefficient or by dividing the actually measured flaw detection data by the correction coefficient.

When the correction coefficient is determined as a ratio of the flaw detection data to the average value, that is, when the correction coefficient=average value/flaw detection data, multiplication is performed. Also, if the correction coefficient is its reciprocal (
Correction coefficient=flaw detection data/average value) is divided to correct the measured flaw detection data.

In the second aspect of the invention, theoretical flaw detection data is obtained in advance for each test position using an equation obtained based on a reflection pattern assuming that an ultrasonic reflecting surface exists on the entire measurement surface of the test material.

Next, the average value of the theoretical flaw detection data at positions where no invalid sound waves and multiple reflections do not occur is determined, and a correction coefficient is determined from the ratio of the average value to the theoretical flaw detection data for each test position. Then, the actually measured flaw detection data obtained by performing a flaw detection test on the test material is multiplied by a correction coefficient, or the flaw detection data is divided by the correction coefficient to correct the actually measured flaw detection data.

[Function] In the invention of claim 1, when a flaw detection test is performed at each predetermined test position on a pair of superimposed reference materials, it is found that there are flaws on the entire superimposed surface (corresponding to the entire measurement surface of the test material). Flaw detection data can be obtained. Originally, the flaw detection data obtained at each test position should be the same, but this flaw detection data is affected by invalid sound waves and multiple reflections. The correction coefficient expresses the degree of influence of invalid sound waves and multiple reflections at each test position, based on a position where no invalid sound waves and multiple reflections occur (usually a central position).

Therefore, by multiplying the actually measured flaw detection data obtained by performing a flaw detection test on the test material by a correction coefficient, or by dividing the actually measured flaw detection data by a correction coefficient, the effects of invalid sound waves and multiple reflections can be removed. According to the invention, therefore:
The accuracy of flaw detection near the edge of the measurement surface of the test material can be greatly improved.

In the invention of claim 2, without actually performing a flaw detection test on the reference material, it is assumed that there is an ultrasonic reflecting surface on the entire measurement surface of the test material (there are flaws on the entire measurement surface), and the reflected echo
Theoretical flaw detection data is obtained for each test position using the theoretical calculation formula obtained based on the pattern, and the correction coefficient is calculated. In this way, there is no need to prepare a reference material, and the flaw detection accuracy can be improved simply by performing a flaw detection test on the test material.

[Example] Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows an example of the configuration of an apparatus for carrying out the method of the invention according to claim 1. In reality, calculations will be performed using a microcomputer. For ease of understanding, an example will be described in which a flaw detection test is performed on the condition of the joint surface of a pair of joined bodies of silicon nitride ceramics. FIG. 2(A) shows the relationship between a pair of ceramics 20a and 20b used as the reference material 20 and the probe 3.
FIG. 2(B) shows the relationship between the ceramic bonded body 20 serving as the test material and the probe 3. The pair of ceramics 20a and 20b constituting the reference material 20 is made of the same material and formed in the same shape as the pair of ceramics 2a and 2b constituting the ceramic bonded body 2. Incidentally, the width W of one ceramic piece is 15 mm, and the height h is 20 mm. The distance between the overlapping surface (measurement surface) 8 and the probe 3 is determined so that the focal point of the focus type probe 3 is located on the surface 8.

When performing a flaw detection test, the probe 3 is moved by a robot capable of moving in the X-Y-Z directions according to a movement pattern as shown in FIG. In this example, test positions are set at intervals of 0.1 mm, and a flaw detection test is performed for each test position.

Note that this interval is arbitrary, and in order to further improve accuracy, it is possible to narrow the interval to 0.01 mm.

The flaw detection data (sum of reflected waves, that is, reflected echoes) measured at each test position by moving the probe 3 is collected by the amplifier Am.
It is amplified by p. Note that this amplifier Amp is built in the ultrasonic pulser/receiver 4 shown in FIG.

The output of the amplifier Amp is converted into a digital signal by the A/D converter 5.

First, a flaw detection test is performed on the reference material 20 prepared as shown in FIG. 2(A) at intervals of 0.1 mm. The flaw detection data for each test position is sequentially stored in a reference material flaw detection data memory 10 having a memory configuration according to a data storage matrix table as shown in FIG. After the flaw detection test for all test positions is completed, the average value calculation means 1
1 to find the average value (hereinafter simply referred to as the average value) of the flaw detection data at positions where no invalid sound waves or multiple reflections occur.

In this example, the average value of the flaw detection data of test positions located within a 3 mm x 3 mm area of the central portion is calculated. Next, the correction coefficient calculating means 12 calculates a correction coefficient for each test position using the average value and the flaw detection data for each test position using the following equation (1).

Correction coefficient = average value/flaw detection data at test position...(1)
The correction coefficients at each test position determined by the correction coefficient calculation means 12 are stored in a correction coefficient memory 13 having a memory configuration according to the matrix class table shown in FIG.

Next, a flaw detection test is performed on the test material 20 shown in FIG. 2(B). In this flaw detection test as well, the flaw detection test is performed using the same flaw detection pattern as when testing the reference material 20, and the measured flaw detection data at each test position is stored in the test material flaw detection data memory 14. The data correction calculation means 15 multiplies the measured flaw detection data of each test position stored in the test material flaw detection data memory 14 by the correction coefficient of each corresponding test position stored in the correction coefficient memory 13. The correction data is output to the display means 16. Note that when the correction coefficient is the reciprocal of the above (1), the measured flaw detection data may be divided by the correction coefficient.

Although the display mode on the display means 16 is arbitrary, the test results can also be displayed by image processing.

FIG. 5(A) shows image-processed flaw detection data (data stored in the reference material flaw detection data memory) obtained when the reference material 20 was subjected to a flaw detection test using a focus type probe. In this image, the ■ part shows the state of the edge of the measurement surface (overlapping surface of the reference material), the ■ part shows the part affected by multiple reflections, and the ■ part is invalid. It shows a part without sound waves and multiple reflections. Fifth
FIG. 5(B) shows an image obtained by multiplying the image of FIG. 5(A) by a correction coefficient to correct reflected echoes. Figure 5 (C)
is an image obtained by image processing the flaw detection data stored in the test material flaw detection data memory 14.

As can be seen from the tendency of the image in FIG. 5(A), even if there is a scratch near the end of the measurement surface, the reflected echo becomes smaller. Therefore, it is difficult to find out whether or not there are scratches around the edges just by looking at the image in FIG. 5(C). Figure 5 CD
) shows an image obtained by image processing the flaw detection data corrected from FIG. 5(C) by the data correction calculating means 15. In the same figure, the part marked ■ indicates a defective part with scratches. If correction is performed using a correction coefficient, the effects of invalid sound waves and multiple reflections at the ends of the measurement surface can be removed, so defects can be reliably detected.

The above embodiment can be similarly applied to the case where a flat type probe is used. Incidentally, FIG. 5(E) is an image of reflected echoes when a flaw detection test was conducted on the reference material 20 using a flat type probe, and no correction was performed. As can be seen from this figure, even when a flood type probe is used, the reflected echo near the end of the measurement surface is smaller than in other parts. Also, the echo from the edge to the center is larger than the center due to multiple reflections. Even in this case, if the present invention is used, the effects of invalid sound waves and multiple reflections can be removed.

The above embodiment is an example in which the present invention is applied to a flaw detection test of a pair of joined bodies of ceramics, but the present invention can also be used for flaw detection of a solid integral body having no joints. In this case, the reference materials may be selected such that the overlapping surfaces of the pair of reference materials are located at a position corresponding to the measurement surface of the solid test material. Other points are the same as in the above embodiment.

Next, an embodiment of the invention of claim 2 will be described.

In the above embodiment, in order to obtain the correction coefficient, it was necessary to prepare a pair of reference materials and perform a flaw detection test on the reference materials. The invention according to claim 2 is characterized in that the correction coefficient is determined theoretically. First, as shown in FIG. 6, assuming that there is an ultrasonic reflecting surface on the entire measurement surface 8 of the test material 2,
Based on the expected reflected echo pattern, the total sum of reflected echoes reflected by the probe is determined as theoretical flaw detection data for each test position. When formulating a theoretical equation, first find the refraction angle and transmittance when the ultrasonic wave hits a solid from a liquid (cases ■ and ■ in Figure 6). In this case, it can be determined using equations (2) to 5) of equation A group below. The symbols used in these equations indicate the values shown in Figure 7, where ρ is the density of the liquid, C is the liquid sound velocity, ρ1 is the density of the solid, C1 is the transverse sound wave of the solid, and CJ
2 is the longitudinal sound velocity of the solid.

(Equation A group) Repeat calculations based on this.

Next, the reflectance of the ultrasonic wave traveling from the solid to the liquid is calculated using the reflectance and refractive index calculated using equation A group.
Calculate using (9) and (10) (■ in Figure 6,
■ and ■). If the reflection occurs multiple times, the calculation results before reflection are used for each reflection, and then the refraction angle and transmittance when the ultrasonic wave traveling with the reflectance obtained from the above calculation hits the liquid from the solid. , (8
) and equation (11) (in the case of ■ and ■ in Figure 6).

Next, among the ultrasonic waves that return to the liquid with the refractive index and transmittance obtained from the above calculation results, the sum of the ultrasonic waves that return to the probe is calculated and used as a reflected echo. In this way, reflected echoes are obtained for each test position and used as theoretical flaw detection data. Then, find the average value of the theoretical flaw detection data at positions where no invalid sound waves and multiple reflections occur, find the correction coefficient from the ratio of the average value and the theoretical flaw detection data for each test position, and store this in the correction coefficient memory. I'll keep it. The subsequent flaw detection test of the test material and the correction of the measured flaw detection data are the same as in the embodiment of the invention of claim 1, and will therefore be omitted.

[Effect of the invention] According to the invention of claim 1, a correction coefficient for removing the influence of invalid sound waves and multiple reflections is obtained using a reference material, and the measured flaw detection data is corrected with this correction coefficient, so that the test The accuracy of flaw detection near the edges of the measurement surface of the material can be greatly improved.

According to the invention of claim 2, since the correction coefficient is determined theoretically, there is no need to actually measure the reference material to determine the correction coefficient, and flaw detection accuracy is improved by simply performing a flaw detection test on the test material. There is an advantage that it can be done.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a schematic configuration of an example of an apparatus used for carrying out the method of the invention of claim 1, and FIG. 2(A)
is an explanatory diagram for explaining the conditions when performing a flaw detection test on a reference material, Figure 2 (B) is an explanatory diagram for explaining the conditions when performing a flaw detection test on a test material, and Figure 3 is an explanatory diagram for explaining the conditions when performing a flaw detection test on a test material. FIG. 4 is a diagram showing an example of a memory matrix table, and FIG. 5 (A) to (
E) is a diagram for explaining image processing results, FIG. 6 is a diagram for explaining reflection patterns, FIG. 7 is a diagram for explaining symbols used in equations (2) to 11), and FIG. Figure 8 is a configuration diagram when performing ultrasonic flaw detection using the water immersion method in general, and Figure 9 is for explaining the reflection state when performing flaw detection using a focus type probe. and FIG. 10 are diagrams for explaining the reflection state when performing a flaw detection test using a flat type probe. 1... Container, 2... Test material, 3... Probe, 4...
...Ultrasonic pulser receiver, 5...A/D converter, 6...Microcomputer, 7...Cathode ray tube, 8...Measurement surface, 10...Reference material flaw detection data memory, 11. ...Average value calculation means, 12.. Correction coefficient calculation means, 13.. Correction coefficient memory, 14.. Test material flaw detection data memory, 15.. Data correction calculation means, 16.. Display means. , 20...Reference material. (A) (B) Figure 5 (C) (D) Figure ■ (E) Procedural amendment (voluntary). February 15, 1990 (2) On page 15, line 2, ``transverse wave sound wave,'' is corrected to ``transverse wave sound velocity,''. (Mu) Patent Application No. 1-342396 2゜Name of the invention Ultrasonic flaw detection method 3゜Relationship with the case of the person making the amendment

Claims (2)

[Claims] (1) In the ultrasonic flaw detection method in which a test material is tested by water immersion method, the test material is made of the same material as the test material, has substantially the same shape as the test material when overlapped, and has a superposition surface that is the test material. Prepare a pair of reference materials that are formed at positions corresponding to the measurement surfaces of , conduct a flaw detection test at each predetermined test position with the pair of reference materials overlapped, and measure the flaw detection data at each test position. , find the average value of the flaw detection data at positions where no invalid sound waves and multiple reflections occur, find a correction coefficient from the ratio of the average value and the flaw detection data for each of the test positions, and perform a flaw detection test on the test material. An ultrasonic flaw detection method characterized in that the actually measured flaw detection data is corrected by multiplying the actually measured flaw detection data obtained by the correction coefficient or by dividing the actually measured flaw detection data by the correction coefficient. (2) In the ultrasonic flaw detection method in which flaws are detected on the test material using the water immersion method, it is assumed that there is an ultrasonic reflecting surface on the entire measurement surface of the test material, and an equation determined based on the reflection pattern is used for each test position. Obtain theoretical flaw detection data in advance, find the average value of the theoretical flaw detection data at positions where no invalid sound waves and multiple reflections occur, calculate a correction coefficient from the ratio of the average value to the theoretical flaw detection data for each of the test positions, and perform the test. An ultrasonic flaw detection method characterized in that the actually measured flaw detection data obtained by conducting a flaw detection test on the material is multiplied by the correction coefficient or the flaw detection data is divided by the correction coefficient to correct the measured flaw detection data.